This may be premature, because some of the the foundational ideas are pretty new and haven’t been thoroughly debated yet, but I thought I’d throw out the bones of a CP, Aggregate and Player Rank overhaul proposal so people can evaluate and critique a big picture approach..

Foundational:

Championship Point Proposal

Calculate CPs by taking a player’s board rating and subtracting 1000. Negative ratings are thrown out, so no score is generated from a rating below 1000.

Using the formula: Standard CPs + (Standard CPs * Bonus %), bonus CPs can then be added to a board's top 10 players' standard CPs based on the following schedule..

#1 = 100% of standard Score, #2 = 90%, #3 = 80%, etc..

Examples

• A player with a 1700 rating on a board receives 1799 - 1000 = 700 CPs

• The #4 player on a board with 1700 receives. 700 + (700*0.7) = 700 + 490 = 1190 CPs

More Examples:

http://www.wargear.net/wiki/doku.php?id=general_features:championship_points_based_on_board_points

All Categories of Play produce independent GRs and CPs

• The categories are: Team Play, Tournament Play, Public Games, and Tournament Team Play.

• If desired, an Aggregate can be generated by finding the mean of all of the above.

Note that it is really not necessary to include GRs in the calculation of an aggregate, because CPs are so closely aligned with GRs in the first place.

Ranks for each category are determined by dividing the range of scores into equal parts.

Rank                                  Range           Example Invention GRs   # of Players (Invention Only)

Brigadier General          Top 1/7 GR               1953 - 2112                                 3

Colonel                         Next 1/7 GR             1792 - 1952                                 1

Lieutenant Colonel                â€¦                     1635 - 1792                                 5

Major                                    â€¦                    1477 - 1634                                 4

Captain                                â€¦                     1319 - 1476                                 8

Lieutenant                   Next 1/7 GR            1160 - 1318                                18

Warrant Officer            Bottom 1/7 GR           1001 - 1159                                109

Private                                                               < 1001

As can be seen in the chart above, a Rank can be determined on a Board Basis, A Category CP Basis, or an Aggregate Basis Across Categories.  This proposal suggests that a Player’s overall Rank is NOT determined on an aggregate basis, though an Aggregate CP score could be tallied.  

Rank is Determined by Highest # of CPs in any one Category

The Rank of 1 Star (Brigadier) General is the highest rank achievable when a player’s CP’s are in the top 1/7 of the range (from top to 1000) of only one Category.  If the rank of Brigadier General is achieved in two categories (e.g. Tournament Play and Public Play), that player is promoted to a 2-Star (Major) General.  3 categories = 3 stars.

The only way to become a 5 star General would be to be a General in ALL four categories, and if more than one person does this, the rank (which is normally reserved for only one person) would go to the highest mean (the proposed Aggregate).

So - hypothetical:

I'm brand new to Wargear, I've been playing risk my whole life. I'm not too bad at it, if inexperienced with the eccentricities of doing this online.

I play a 6 player game of  WGWF with the following GR scored players: 989, 1050, 1267, 826 and 1172

I win. 

My GR goes up to 1081

I get 81 CP for my first win?

I like this system, but it seems cumbersome if on the top end players will be having like 10,000 cp's total and I'm getting 80 for my first single win. 

Is there anyway we can scale it back to nicer, smaller, cleaner numbers?  So I get like 1 for my first win, and have a total someplace around 1000-3000?  Or is that too close to the GR numbers calculation?  Could we divide by 1000 instead of subtract?  But isn't this just adding together all your GR on all the boards?

ratsy wrote:

hypothetical:

yes

I like this system, but it seems cumbersome if on the top end players will be having like 10,000 cp's total and I'm getting 80 for my first single win. 

The road is long and arduous, but at least you're on it and you have a clear sense of what it takes to rise in the rankings - unlike the current system where you can play and play and not get a single CP.

I did a quick calculation using berickf and itsnotatumor's game data, and their current CPs under this system come out to about 7600 and 15,680 respectively.  No doubt berickfs CPs would be high in other proposed CP categories, but that data doesn't currently exist.

Is there anyway we can scale it back to nicer, smaller, cleaner numbers?  So I get like 1 for my first win, and have a total someplace around 1000-3000?  

1 on the way to 1000 is the same getting your first 20 on your way to 20,000.  Sure, you could divide by 1000 and bring the calculations out a few decimal places, but then the meaning gets lost.  The whole point of this system is that CPs are directly derived from game points.  You can look at the field and pretty much estimate right as you start a game how winning and losing will affect your CPs.  An 80 point win gets you 80 CPs.

Right now we're all conditioned to see CPs going up to a few hundred or so, but give the proposed scale a little time and I'll bet it will become as natural as anything else.  Besides, and especially for those who most highly value the concept of CPs, doesn't it make sense that they should be the largest numbers?

 But isn't this just adding together all your GR on all the boards?

Almost.  The difference is that GR scores less than 1000 on any board are thrown out when calculating CPs.  CPs only apply to boards you "win" at.  Of course, we could still keep the old school GR scores (which would include ALL board scores), and technically you could include these in an Aggregate for each category, but that would be confusing, inelegant and somewhat redundant given that CPs are so closely aligned to GRs in this system.  Besides, who wants to know how "bad" they are?

Also, don't forget that there's an additional scaled bonus point incentive for getting into the top 10 on each board, where the top player gets double CPs.

I hope to figure out the scores for a few more and post some numbers, but in the case of the two I've already done.

I'm crunching some numbers and I've come across a problem (I think) with the Ranking part of the Proposal. The problem has a name, and that name is Cona Chris.

He's so dominant  (#1 on 17 boards) that he'd be the only General in this category. A bump in rank requires 4919 points (1/7th of 34,430), and I wonder that Mad Bomber could even make Colonel at that scale.

As you all know I'm not a fan of arbitrary delineations based on magic numbers, so I'm somewhat stumped here. I don't want 1/7 of the players to be generals, 1/7 to be colonels, etc.. If anything, things should be very steep at the top.  Fewer than 1/2 of 1% of commissioned officers reach any of the General Ranks in RL.

One idea that comes to mind is to mandate a certain % of generals - Maybe something like the top 2% - then Scale everything evenly there by score (1/6ths).  Yes 2% is arbitrary, and the system isn't nearly as elegant, but that's all I've got right now.